Document of

The World Bank

FOR OFFICIAL USE ONLY

Cyprus at Crossroads:

A Public Sector for the Post-Crisis Economy

FINAL REPORT

April 30, 2014

This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.

Table of Contents

PrefaceiiExecutive SummaryiiiA.Cyprus at a Crossroads1B.A More Flexible, Performance-Oriented HRM System2 The Patronage System and its Consequences2 Limiting Rusfeti and Enhancing Managerial Discretion: Reform Options6 Reform Options for Enhancing Managerial Responsibility8 Complementary Reform Options for Greater Effectiveness10 Reform Options for the Performance Appraisal Process11 Strengthening Performance in Promotion Decisions13C.Pay Policies Not Linked to Performance14 Reform combined salary scales20 Reform “annual increments”22 Undertake a phased, selective pay freeze approach to address imbalances23D.An Affordable Salary Structure24E.A More Accountable and Transparent Public Service29F.An Integrated Approach to Reform31G.References35Annex 1: Data and Methodology42Annex 2. Lessons from Performance Appraisal Systems in the EU46Annex 3: Career versus Position HRM Systems49Annex 4: International Experience with Pay for Performance50Annex 5. Supplementary Wage Bill Analysis52Annex 6: A Review of Working Conditions in the Public Sector56Annual vacation leave56Sick Leave57Maternity Leave59Working Time60

60

Preface

This report was produced by a team comprising staff from the World Bank (WB) in partnership with the United Kingdom’s National School of Government International (NSGI) at the request of the Government of Cyprus. Three WB and NSGI missions visited Nicosia from May 20-24, June 25-28, and September 9-20, 2013, to agree to the terms of reference and costing, and to undertake the analysis and meet with key stakeholders. This version incorporates the comments received after an earlier version was presented to Government officials during the week of December 9th, 2013

The team is extremely grateful to Ms. Emmanuela Lambrianides (the Commissioner for Public Sector Reform), Mr. Christos Patsalides (Permanent Secretary, Ministry of Finance), officials from the Public Administration and Personnel Department, and the many other Cypriot officials for their contributions and hospitality. The team also wishes to acknowledge the contributions and assistance from external stakeholders, including representatives of the public sector unions, academics, civil society, and the independent members of the Steering Committee.

The primary objective of this report is to provide an objective assessment of human resource management policies, institutions and practices with proposals to improve the effectiveness, efficiency, and responsiveness of the public sector to deliver better services to its citizens. Throughout this report, the analysis is confined to the core public sector and excludes employees of State-Owned Enterprises. This report is a draft, subject to further discussions with the Cypriot authorities, as was agreed in the terms of references.

Executive Summary

The Cyprus public administration is set to embark on a change process. An important element of the Government of Cyprus’s (GoC) economic adjustment program under Troika financing is public administration reform. The objective of these reforms is to ensure the efficient use of government resources, while delivering a quality service to the population. The World Bank Group in collaboration with the UK’s National School of Government International (NSGI) has been tasked with providing technical advice to the GoC on reform options to help enhance effectiveness and efficiency of the public sector. The primary objective of this work is to support the GoC in modernizing, streamlining, and strengthening the performance of the public administration.

Improving the effectiveness and efficiency of service delivery will be essential in the post-crisis environment. While the crisis may have provided the impetus to re-think the role and function of the public sector, public administration reform in Cyprus is long overdue. Cyprus has not undergone any significant public sector reform in the last four decades. Consequently, its public service is ill-equipped to manage the changing dynamics of the population and the economy, which will have to shift from a reliance on financial services to other sectors to drive growth in the medium-term. The main findings of this report, therefore, are not only proposed technical solutions for modernizing the public sector they can also be considered as an integral element of the GoC’s emerging growth strategy. The modernization of the public service should also support other public sector reforms, in particular reforms in planning and budget systems which will determine the allocation and application of resources across the public sector.

The Cypriot public service is bound by a rigid, anachronistic system of laws and regulations that hinder effective personnel and performance management. Policies have been adopted over the years to address the risks posed by rusfeti, a patronage system that undermines merit based selection, pay and promotion. Unfortunately, these measures have had the unintended consequence of making the Human Resource Management (HRM) system excessively rule-bound, centralized, and inflexible. Undue reliance is placed on seniority as the main determinant of promotion. The annual personnel performance appraisal process fails to distinguish between staff. Virtually all HRM decisions involve the Public Service Commission (PSC), which leaves managers with limited discretion for HR functions, hinders performance management, and compromises efforts to hold managers accountable for the performance of their organizational units.

There is no effective link between pay and performance in the public service. Individual salary increases and the overall wage bill are driven by a combination of annual increments, compound increases, and cost of living adjustments (CoLA), which accrue to every civil servant each year regardless of performance. Combined salary scales allow public servants to move up the incremental steps across multiple grades without having to be promoted, guaranteeing automatic annual pay increases throughout an individual’s entire career (18-29 years). This policy does little to motivate staff, contributes to relatively high salaries (particularly late in staffs’ careers), and leads to a very weak relationship between the human capital requirements of positions, their levels of responsibility and remuneration.

More Managerial Discretion, More Managerial Accountability and Affordable

The recommendations presented herein represent a move towards introducing more flexibility into the HRM system. The move to more managerial discretion in HRM decisions does raise the risk that patronage may increase under the new system, but our suggested changes to personnel decisions and the architecture of the grievance redress mechanism are meant to curtail this possibility, while at the same time reducing the burden on the Supreme Court. Similarly, there are a number of suggestions to protect the integrity of managerial decisions related to recruitment, performance assessment and promotion. These recommendations imply a shift in the balance between rules and discretion in the HRM system, which can be further adjusted as lessons are learned from implementation.

Measures to enhance managerial discretion go hand in hand with measures to enhance managerial accountability. The Fiscal Responsibility and Budget Systems Law provides a foundation to create managerial incentives to cost-effectively meet the policy and program objectives. The Law puts in place a budget and medium-term expenditure planning process that establishes a hard budget constraint for the public sector as a whole and individual institutions, and requires specification of organizational units’ and program objectives. It also encourages management to consider the tradeoffs between the application of scarce resources on wages and salaries, capital investment, and operations and maintenance in terms of their contribution to organizational and program objectives, which focuses attention on efficiency and effectiveness. Furthermore, it creates a framework for organizational accountability, since progress towards the achievement of objectives can be monitored and assessed by policy makers, the legislature and the public.

A more effective public administration must be affordable. Organizational reforms to the HRM system will show long lasting improvement only if they are affordable. A key recommendation that both enhances effectiveness of the public administration and has the additional benefit of containing future wage bill growth is to reform the salary regime. The suggested reforms in the salary regime aim to strengthen the link between remuneration and performance. Again the system has to achieve a balance between potentially conflicting objectives. Fiscal sustainability, efficiency and value-for-money considerations will tend to constrain salary growth and keep salaries at or below market rates. The need to recruit experienced, competent staff with scarce skills, retain these staff and ensure that they are properly motivated will tend to promote higher levels of pay at or above market rates. The suggestions below on pay reform aim to reconcile these objectives.

A Package of Mutually Reinforcing Reforms

Organizational reforms to the public service must decentralize responsibility for human resource management while ensuring adequate oversight. Three changes in the organizational arrangements are proposed:

· Responsibility for human resource management should be devolved to line ministries and agencies. Decisions related to recruitment, performance assessment and promotion should be taken by the line managers responsible for organizational performance. However, within the line agency, no single individual should be responsible for a HRM decision. Successive stages of the decision making process should be assigned to different individuals to ensure that these decisions are reviewed and contested. In the case of the recruitment process, for instance, decisions regarding long-listing and short-listing final selection would be taken by committees and the final selection decision taken by a line manager.

· The PSC should assume an oversight role. The PSC’s key functions should include: communicating policy, verifying compliance with HRM procedures, and assuming responsibility for the administrative review of specific HRM actions where these are contested. In order to fulfil this oversight role credibly, the PSC has to be constituted as an independent entity, with its members appointed through a transparent process on the basis of professional competence, ideally comprising representatives from outside the public service and without political affiliation.

· The architecture of a grievance redress system has to be put in place. This grievance redress system should offer options for individuals to contest HRM decisions along a continuum of formality, including peer support (more informal), an administrative ombudsman, mediation, an administrative tribunal, to a direct appeal to the existing Supreme Court judicial review process (more formal). The objective is to create a set of less confrontational, lower transaction-cost means of addressing perceived unfairness and complaints regarding the HRM system and HRM decisions.

These organizational changes seek to provide managers with greater flexibility in human resource management by shifting from ex-ante to ex-post external controls. This shift is consistent with the trend in public sector management in most OECD countries over the last three decades. This strategy is not without risks, however, as management discretion is open to abuse. The dismantling of controls that sought to limit the scope for rusfeti could lead to the resurgence of this practice, which would undermine the very purpose of reform. Close monitoring of progress in the implementation of these reforms – taking into account the perception of those within and outside of the public service – will help mitigate these risks. The authorities may also wish to consider making the organizational reforms conditional and leave open the option to restore external ex-ante controls where agencies are unable to put in place an adequate internal control framework. With implementation, adjustments can be made as lessons are learned and where progress is made, further relaxation in ex-ante controls may be considered.

A second set of changes in the HRM system are primarily geared to creating incentives for improved performance. The key elements of such a system are merit-based recruitment and promotion, a salary structure that is linked to performance rather than seniority, and arrangements that foster mobility within the public service:

· The annual personnel performance appraisal process needs to be strengthened to ensure that merit and performance are the key determinants of promotions and salary growth. The current proposal from the Public Administration and Personnel Department (PAPD) is a welcome initiative and, encouragingly, builds in a staff development component for managers to provide more continuous feedback on performance. It is, however, missing one important feature. Currently, there is no device to discipline the pattern of ratings assigned to staff, either at the ministry or agency level. This recommendation was initially included in an earlier proposal, but was subsequently dropped due to opposition from various stakeholders. A new performance appraisal system is being developed in cooperation with the UK’s National School of Government International (NSGI) and it has a disciplining mechanism for ratings as a core feature.

· Changes to the salary structure and its dynamics will reinforce efforts to decentralize responsibility for HRM functions. There are three complementary reforms to the pay regime that are being proposed that would strengthen the link between remuneration and performance:

· Restructure the combined salary scales;

· Limit the value of automatic annual increments in base pay or, better still, eliminate automatic annual increments and introduce performance-related bonuses for staff that perform well or consolidate annual increments;

· Link increases in public sector pay to comparators in the private sector based upon a robust and independent analysis of market conditions for specific job types, taking into account the need for decompression of the salary scale to provide competitive salaries for competent, experienced staff with scarce skills.

· The HRM system should encourage staff mobility across the public administration so that staff can be redeployed to meet changing business needs and provide staff with richer more rewarding work experience. While there have been some recent changes to the process governing secondments, barriers, such as unduly restrictive schemes of services, continue to restrict mobility. At the same time, there are no career-enhancing incentives for individuals to move around the public administration. Many of the reforms proposed here (increasing open competitions for mid-level posts, stricter performance management, reducing the number of steps in the combined salary scales, limiting the number of annual increments, and reducing the numbers of schemes of service) should, over time, encourage public employees to see mobility as both positive and career-enhancing. Experience gained in more than one Department could be made a requirement or advantage for promotion.

An Enabling Environment for Reform

Technical solutions must be complemented by a change in organizational culture. The proposed technical solutions above will only fully deliver the anticipated benefits if there are associated reforms that change organizational culture, reinforce ethical behavior and clarify expected behaviors, roles, and responsibilities of public actors. Additional reform options are presented to reinforce the adherence to the new code of ethics, and increase the understanding of the roles and boundaries among public servants, politicians, and political appointees.

Addressing the challenges identified in this report will require an integrated reform effort to bring about effective change. Many of the proposed reforms will have little impact if they are undertaken in isolation or not properly sequenced. For example, the weak linkages between performance and promotions prospects cannot be addressed without first finding a reliable means of objectively assessing individual performance. In addition, the incentive for managers to assess and motivate staff will be higher if they are held accountable to achieving their unit’s performance targets.

To implement these reforms, the Government will need a change management strategy and to communicate clearly and frequently to the public servants and the public. Public administration reforms are never easy and are rarely popular. Therefore, it is imperative that the Government build broad support beyond the public service – among politicians, the business community, citizens who are the taxpayers and beneficiaries of public services – to generate positive momentum for reform. In addition, although there is a large constituency in the current system that benefits from the status quo, there are also constituencies that can be tapped for support.

Cyprus at a Crossroads

Cyprus is in the aftermath of a banking sector crisis. The Cypriot authorities agreed to the terms of a €10 billion bailout program with the Troika in early April 2013. The banking system amassed large holdings of Greek sovereign debt and was hard hit by the Greek sovereign credit write-down during the Eurozone crisis. Non-performing loans at the nation’s three largest banks rose to more than 40 percent in early 2014. As the banking sector had grown to almost 9 times GDP before the crisis, recapitalizing the banking sector has important fiscal consequences. The Troika’s financing alone represents more than 50 percent of 2013 GDP, which immediately increased the public debt to 112 percent of GDP. This level of public debt and the ongoing contraction in output implies fiscal sustainability is a key concern going forward.

The banking crisis presents an opportunity to revisit the efficiency and effectiveness of the public administration. Anchored by EU accession in 2004 and the euro adoption in 2008, primary fiscal surpluses were the norm in the run-up to the crisis. Cyprus experienced above EU average real GDP growth rates of close to 4 percent during the period as the financial sector and ancillary services took off. Nevertheless, a compositional change in the fiscal accounts took place over this period. The public wage bill grew to the largest in the Eurozone and remuneration levels are substantially higher than the EU average, crowding out expenditures on other necessary inputs, such as investments, social protection and non-wage operating costs. Consequently, an important element of the Government of Cyprus’s (GoC) economic adjustment program under Troika financing is public administration reform. The objective of these reforms is to ensure the efficient use of government resources, while improving the quality of service delivery to the population.

A modern and more efficient public sector is a worthy objective in its own right. Fiscal savings should not be the sole motivation for public sector reform in Cyprus. Indeed, international experience (World Bank 2008) suggests that the drive for short-term fiscal savings does not lead to lasting public administration reform. Rather, reform efforts need to be driven with the aim to improve the effectiveness, efficiency, and responsiveness of service delivery for the population. Cyprus has not undergone any significant public sector reform in the last four decades. Consequently, its public service is ill-equipped to manage the changing dynamics of the population and the economy, which will have to shift from a reliance on financial services to other sectors to drive growth.

Crisis measures to contain the growth in the wage bill risk being short-lived if not supplemented with institutional reforms. A number of crisis policy measures will help bring wage bill expenditures more in line with EU averages and address some of the concerns about the cost of the wage bill. One key measure is an earnings freeze for public sector workers and pensioners until December 2016 as well as a freeze in the cost of living allowance (CoLA) adjustments until the end of March 2016. While these and the other proposed measures will help contain growth of the wage bill until early 2016, fundamental changes to pay and employment policies will be needed to ensure a more effective and efficient public administration that is affordable over the medium-term. The objective of this report is to provide a comprehensive set of institutional and organizational reform options that create incentives for improved performance and decentralize responsibility for human resource management while ensuring adequate oversight.

A. A More Flexible, Performance-Oriented HRM System

Key Findings

· A patronage system known as rusfeti is the most significant challenge in reforming HRM practices

· Lack of trust in the HRM system to deliver fairness and reward for merit and high performance has led to many of the current restrictive and inefficient HRM practices

· Staff rely on judicial avenues in order to seek redress for HRM decisions

· Managers lack the authority and means to take critical HRM actions

Proposals

· Ensure that no major HRM personnel action can be controlled by any single agent

· Reduce the odds of a “captured” Public Service Commission

· Ensure greater contestability in major HRM actions

· Create an independent redress mechanism for HRM disputes

The Cypriot public service is bound by a rigid, anachronistic system of rules, laws, and regulations that aims to limit patronage. These traits lead to an inflexible system where the incentive structures are misaligned. HRM policies discourage and hinder staff mobility and high performance, which limit the ability of managers to manage staff and achieve efficiency gains and better outcomes. This section presents the key features of the HRM system and suggests a change in the orientation of the public sector to one in which merit and performance are the key determinants of promotions, salary increases, and career prospects. A key consequence of these reform options is the introduction of more managerial discretion in HRM decision making, possibly resulting in to the resurgence of decisions made on the basis of patronage. Our suggested changes to the architecture of the grievance redress mechanism are meant to curtail this possibility, while at the same time reducing the burden on the Supreme Court.

The Patronage System and its Consequences

Patronage seriously compromises the integrity of the Cyprus HRM system. In Cyprus, a particular form of patronage known as rusfeti is widely recognized as being a fundamental determinant of many major personnel actions, such as appointments, promotions, and transfers.[footnoteRef:1] Rusfeti depends on a mutual sense of interpersonal obligation, largely reflecting a history of reciprocal personal favors, provided when needed rather than as a single transaction quid pro quo or exchange of money for the favor.[footnoteRef:2] Having a “patron” is more important factor in determining career progression than being either well qualified or very productive. This practice demoralizes hardworking and high performing staff since less productive staff may be promoting ahead of them due to better connections. [1: Faustmann (2010), p. 270. The single important exception to this appears to be selection to entry level posts, in which the required written examination is both sufficiently rigorous and a significant enough consideration that virtually all candidates making it to the short-listing stage are recognized to be well qualified.] [2: Faustmann (2010), pp. 269-289.]

Because rusfeti is driven both by patrons and their clients, it is extremely difficult to reduce or eliminate. Since the practice is pervasive, actions aimed at reducing it will unavoidably meet with resistance. Consequently, unilateral action against rusfeti by individuals is ineffective and discouraged by the threat of negative consequences. Although an anti-rusfeti law was enacted in 2001, making rusfeti a punishable offense, only one case had been filed and no case has been successfully adjudicated to our knowledge.[footnoteRef:3] [3: Faustmann (2010).]

The indirect impacts of rusfeti have had a more significant effect on HRM policies and practices. In particular, a number of the policies that have been adopted over the years to address the risks posed by rusfeti have had the unintended consequence of making the HRM system too centralized and inflexible, which has undermined HRM efforts to enhance performance of both organizational units and individual public servants. For example, it has led to:

· Undue reliance on seniority as the main determinant of promotion despite its poor correlation with either ability or performance;

· The inability of the annual performance appraisal process to actually rank staff by performance;

· A near monopolization of major HRM decision authority within the Public Service Commission; and

· Limited managerial discretion, which undermines managerial capacity to manage for performance and compromises any attempts to hold managers accountable for the performance of their organizational units.

Current HRM practices do not facilitate the application of a meritocratic principle. Under current law, promotions decisions are supposed to be based upon three criteria: qualifications, performance, and seniority. However, almost all staff meet the minimum criteria required in the scheme of service.[footnoteRef:4] Staff can differentiate themselves by securing additional academic qualifications. Promotion decisions do not, however, distinguish between differences in the quality of qualifications, nor do assessments provide any other evidence on a candidate’s skills, knowledge or other work-relevant traits. This policy creates incentives for staff to seek educational credentials that may not be relevant for their employment. Assessments also do not provide for systematic testing of particular skills or knowledge that may be important in a particular career stream, such as interpersonal skills, teamwork, analytic skills, and leadership skills. Moreover, except for entry level positions, formal testing is neither required nor practiced. [4: Schemes of service provide detailed specification of job requirements, duties and responsibilities and terms and conditions of employment for a particular position within a given career stream, identifying for the covered position: (a) salary scale applicable to positions within the scheme of service; (b) duties and responsibilities; and (c) qualifications required. Salaries are determined through the budget, based on the applicable salary scales and other parameters governing remuneration.]

Performance plays little to no role in promotion decisions. Staff member’s annual performance evaluations do not differentiate between high and low performers because almost all staff are rated as “excellent.” In 2011, for example, 69 percent of staff received “excellent” (the top rating) in all eight performance criteria categories, while 99 percent of staff received at least six “excellent” scores and no negative ratings.[footnoteRef:5] The Supreme Court has ruled that the variance in performance ratings among staff that had 6 to 8 “excellent” scores was too small to warrant differentiating candidates based on measured “performance”.[footnoteRef:6] Since this ruling equalizes 98 percent of staff, performance has almost no impact on promotions decisions. [5: Public Services Commission Report of the Chairman, 2003-2013. These rating patterns are consistent over time. Between 2000-2011 the percentages for these three groupings averaged 71%, 93% and 98%, respectively, and never dipped below 67.8%, 91.4% and 97.6%, respectively.] [6: Interview with PADP staff (September 10, 2013).]

Concerns about rusfeti have also led to the centralization of HRM authority. Constraints have been imposed to limit managerial autonomy due to concerns about rusfeti. Major personnel actions are taken by the Public Service Commission (PSC) leaving HRM authority outside of the PSC severely constrained (see Figure 1). The five members of the PSC have a near monopoly on most day-to-day and important career-determining HRM actions.[footnoteRef:7]. [7: The sole exceptions being that (a) promotions within a Department are the responsibility of the PSC, based upon the recommendation of the Head of Department within which an eligible vacancy exists, and (b) annual personnel performance evaluations are undertaken by each Ministry, under the oversight of the PSC. Art. 98, Constitution of the Republic of Cyprus, as well as The Public Service Laws of Cyprus: 1990-2006 (see, in particular, Art. 5, which enumerates the functions of the PSC, employing the same text as is found in Art. 98 of the Constitution, and fails to identify annual performance appraisals as one of those functions).]

Concerns about rusfeti have also contributed to extremely low managerial discretion for HRM in the Cyprus public administration. A recent review of the PSC concluded that a detailed, rigid legal framework seriously compromises the capacity of managers within Cyprus’ public administration to exercise the sorts of discretion that are typically essential to effective HRM practices.[footnoteRef:8] These constraints are by design to limit the risk of misuse or abuse of such managerial discretion in the presence of rusfeti. Examples include: [8: “The Cyprus Public Service Commission: An Overview”, (commissioned by the PSC, 2003), p. 7: http://www.psc.gov.cy/psc/psc.nsf/0/9B57526179C43BC9C2256FB6003D82A3/$file/An%20overview.pdf?OpenElement ]

· Experience is not allowed to substitute for education, or vice versa in typical “Schemes of Service”.

· Annual performance appraisal forms employ eight identical criteria for all positions, without any provision for tailoring the criteria to the activities required by or performance objectives of a particular position.[footnoteRef:9] [9: The criteria are: (i) professional development; (ii) performance; (iii) interest; (iv) responsibility; (v) initiative; (vi) cooperation; (vii) attitude toward citizens; and (viii) managerial skills.]

· External candidates from outside the public administration are not allowed to compete for openings being filled through the within-Department promotion procedure.[footnoteRef:10] [10: This constraint is typically imposed by the qualifications specified for such posts in their “Scheme of Service”.]

· Relevant experience outside Cyprus’ public administration is not allowed to be considered in within-Department promotions decisions; only internal experience counts.[footnoteRef:11] [11: Again, this constraint is typically imposed in the “Schemes of Service” for a given position.]

· Qualifications are interpreted solely as academic certificates and degrees, with no consideration of the quality of those credentials or the institutions that provide them.

· Seniority is, de facto, the primary determinant of within-Department promotions.

· Informal pressures encourage managers to reward their staff with the highest score in the annual performance ratings.

Table 1. Formal HRM Responsibilities in the Cyprus Public Sector

Sole Authority of PSC

Recruitment, selection, appointment and promotion

· Entry level appointments (“First Entry”)

· Mid-career appointments (“First Entry and Promotions”, from scale A14 and above)

· Within-Department promotions “Promotions”)

Retirements

Redress

· Adjudication of disciplinary actions proposed by Departments, and imposition of any sanctions, including dismissal.[footnoteRef:12] [12: In addition, the Commission has authority to both initiate proceedings and impose sanctions upon public servants for contempt of the Commission or for failure to appear before the Commission when so ordered (The Public Service Laws of Cyprus: 1990-2006, Art. 19).]

Primary but shared authority

Transfers (reassignments that require a change in the place of residence)

· Upon a proposal by “the appropriate authority”, taking into consideration the views of the employee.

Secondments

· Subject to the views of both the receiving and contributing Ministers, and the Minister of Finance who has a deciding role, and the final approval of the PSC.[footnoteRef:13] [13: A recent change in the legal framework removed a third clearance for secondments; namely, by the affected civil servant and also a fourth restriction on the duration of the secondment.]

Oversight Authority (with no direct authority to undertake these particular HRM actions)

Annual personnel performance assessments

· The Public Service Laws are silent on PSC’s function regarding the annual personnel performance review requirements.[footnoteRef:14] The detailed regulations governing performance appraisals are proposed by PAPD, subject to approval by Cabinet and the House of Representatives. The actual appraisals are undertaken by each managing authority (i.e. by individual ministries, departments, and agencies). The Public Service Commission's involvement in the assessment procedure is limited to (i) collecting data on the assessment of each employee each year by the Permanent Secretaries, and (ii) collecting the most recent evaluations of employees to be taken into account during promotion procedures (evaluation of the “merit” criteria).[footnoteRef:15] [14: The Public Service Laws of Cyprus: 1990-2006, Art. 5, which enumerates the functions of the PSC and fails to identify annual performance appraisals as one of those functions, despite the fact that Art. 35(4) mandates that the “Annual Appraisal Reports” of candidates for promotions shall be an important consideration in the Commission’s deliberations on all promotions.] [15: “The current appraisal system, as applied, is neither efficient nor functional at all. A single form is used for all categories of employees, regardless of whether they are physicians, clerks, engineers or they are at the top or lowest hierarchical level of the structure.” Source: “The Cyprus Public Service Commission: An Overview”, page 7.]

In short, promotions decisions are driven by either of two factors, rusfeti or seniority, without any clear relationship to merit or performance. For management positions, rusfeti is widely believed to be the determining factor in the selection process in most cases. For non-management positions, seniority is the determinant factor in the selection process since there is little or no variance in the scores of candidates along the other two criteria for promotions. These practices reward either loyalty (through rusfeti) or internal experience (through seniority), which generate a misaligned incentive structure where merit or performance is not rewarded or encouraged.

Limiting Rusfeti and Enhancing Managerial Discretion: Reform Options

Reducing rusfeti practices and their indirect impacts on HRM policies and practices is the most important and the most difficult challenge facing the Cyprus public sector. The lack of trust in the HRM system to deliver fairness and reward for merit and high performance has driven many of the current restrictive and inefficient HRM practices, which have led to an overreliance on judicial avenues to seek redress. Reform of the HRM system must also therefore include measures to mitigate the underlying causes of this lack of trust. There is no “silver bullet” or reform that will dramatically reduce, let alone eliminate these practices, because they are so deeply embedded in behavior and mores (Table 2 provides a series of international best practices). Nevertheless, there are actions that could, collectively, limit its influence.

Table 2. Mechanisms to Minimize Patronage

Due Process Protections

· Dismissal only for cause: Cause typically includes unsatisfactory performance appraisals, conviction of a felony, or incapacity to perform one’s duties, etc.

· Third party review: Major personnel actions require review by an independent third party.

· Audit trail: Major personnel actions generate an audit trail.

· Independent redress: Major personnel actions can be challenged through a redress mechanism that is independent and credible.

Recruitment and Selection

· Competition requirements: Subject all competitive positions to strict rules and criteria for evaluation candidates, such as interviews, desk reviews of candidates’ submitted materials (CVs, reference letters, writing samples, etc.), formal behavioral assessments (e.g., through role-playing exercises, etc.), professional knowledge, interpersonal skills, teamwork aptitudes, and relevant experience.

· Tiered screens: Multi-staged recruitment and selection procedures (long-listing, short-listing, final selection) in which no single agent is likely to be able to determine outcome of each stage.

· Independent redress: A redress mechanism that is independent and credible, where aggrieved candidates can challenge a particular recruitment and selection decision.

Promotion Procedures and Practices

· Emphasis on performance: Give greater weight to performance than experience or seniority in promotions decisions.

· Mechanisms to discipline promotion practices: Competition requirements, tiered screens, and independent redress (as described above).

Disciplinary Procedures and Practices

· Disciplinary actions commensurate with infractions: Provide a range of disciplinary options and clear guidance to ensure that disciplinary actions are commensurate with the seriousness of the infraction.

· Checks on disciplinary actions: Require that the disciplining agent include an attributable statement to justify the disciplinary action, which is subject to the review and clearance by a superior or another independent agent.

The first step is to ensure that no major personnel action can be controlled by any single agent. In Cyprus, this recommendation would entail breaking the PSC’s monopoly on major HRM actions by devolving some responsibilities (such as recruitment, selection, and promotion) to line ministries and agencies, subject to contestability requirements and checks on the execution of those responsibilities. Although this approach would require a Constitutional amendment, it has been followed by many OECD countries to limit the influence of private interests, which can always be expected to strongly influence the arguments and actions of any given agent, often to the detriment of “the public good”. In the United Kingdom, for example, no key personnel decision (appointment, promotion etc.) is decided by a single individual. Instead, it relies upon an internal system of counter-signature by senior officials and an escalation of complaints and appeals. Once internal avenues have been exhausted, appeals can be taken to an independent arbiter (either specific to the sector or under general employment law). In the case of the Civil Service, the Civil Service Commission acts as the independent regulator of all appointment processes. The emphasis therefore is on internal dispute resolution processes that are robust and enjoy the trust and confidence of all parties.

The second step is to strengthen the independence, transparency and oversight of the PSC. Currently, the chairman and the four other members of the PSC are political appointees for six year terms. This appointment process contributes to the perception that PSC decisions are political in nature or influenced by rusfeti. To minimize the perception of bias, the Government could change the membership of the PSC through an amendment to Article 124 of the Constitution to include not only Presidential appointees, but also members appointed via multiple, competitive, and merit-based processes. These new members could include HRM professionals, representatives of specific professions, or representatives of the civil service, which, because of their differences, would be likely to represent diverse “factions” or interests. Other European countries have followed this example. In the UK, for example, members of the Civil Service Commission (and other public service appointments commissions) are recruited through an open, competitive process, which is designed to select Commissioners from a range of backgrounds and with HRM expertise (public service, private sector, trade unions). The appointments process produces candidates that are recommended on merit to the Prime Minister for appointment. Commissioners are appointed for fixed five year terms, which are staggered to ensure that there is continuity of experience and expertise and that membership spans changes of the governing administration.

The third step is to ensure independent review and contestation of HRM actions. Merit -based selection requires that different entities are responsible for decision making at each stage in the selection process: long-listing, short-listing, and final selection. Long-listing, for example, could be undertaken by professional HRM staff, while short-listing could be undertaken by the PSC; professional HRM staff; professional staff with the same expertise required by the advertised position; and a representative of the entity in which the selected person will be posted. The final selection process should be undertaken by the immediate supervisor of the position to be filled.

The fourth step is to create a set of less confrontational, lower transaction-cost means of addressing grievances against HRM actions. Currently, all HRM decisions that are challenged go directly to the Supreme Court by law, which not only clogs the system, but also causes internal conflict, demotivates staff, and destabilizes the leadership of ministries. Taking grievances directly through a formal judicial adjudication process makes such challenges almost unavoidably confrontational and imposes unnecessarily high transactions costs. The aim of this reform would be to transform that redress process into a less confrontational, less time-consuming process; i.e., into a process that both provides credible redress, while simultaneously reducing the average length of grievance resolution.

While some challenges will unavoidably become confrontational and will require a judicial resolution, Cyprus’ public servants would be better served by having access to a set of less formal grievance resolution mechanisms. These mechanisms should make it easier to resolve grievances definitively through informal, non-confrontational processes, and only send grievances to a formal judicial process if they prove impossible to resolve through those simpler, more informal, less confrontational procedures. A range of options exist along a continuum of formality for addressing this need, from peer support (more informal), to an administrative ombudsman, to mediation, to an administrative tribunal, to a direct appeal to the existing Supreme Court judicial review process (more formal).

The World Bank, in fact, when it reformed its conflict resolution processes several years ago, introduced a Conflict Resolution System (CRS) with all of these elements. That CRS includes: (a) Respectful Workplace Advisers (i.e., peer support); (b) Ombuds Services; (c) Mediation Services; (d) Peer Review Services; (e) a formal Administrative Tribunal (the equivalent of a judicial mechanism within the World Bank).[footnoteRef:16] The basic idea is to provide a range of options for trying to resolve a grievance, and only provide access to a formal judicial mechanism as a last resort. [16: For details of the World Bank’s Conflict Resolution System, see: http://web.worldbank.org/WBSITE/EXTERNAL/EXTABOUTUS/ORGANIZATION/ORGUNITS/EXTCRS/0,,menuPK:64165918~pagePK:64165931~piPK:64166031~theSitePK:465567,00.html ]

Reform Options for Enhancing Managerial Responsibility

The options outlined for enhancing managerial discretion go hand in hand with options to enhance managerial responsibility. The ongoing Medium Term Budgetary Framework (MTBF) reforms and wider strategic management reforms provide an opportunity to create managerial incentives to link HRM with the policy and program objectives assigned to their organizational units, such as ministries, agencies, directorates, and departments. Any MTBF process includes specification of the policy and program objectives of organizational units, as well as the production of regular and timely evidence on whether targets are being achieved. If these objectives are to be realized, performance measures will have to cascade through the organizational structure: from ministries, through departments and units down to the level of individual managers and their staff. If organizations and managers are to be held accountable for performance, they must also be allowed flexibility in the allocation and application of resources. The following reforms could build on this process to by linking the performance of individual staff with their career prospects:

· Hold managers accountable for the performance of their unit;

· Cascade the ministry/agency-level objectives down to lower level organizational units within each ministry/agency and agree on performance indicators;

· Establish systems and capacity for systematically and continuously monitoring performance indicators;

· Build organizational unit capacity to use this data to inform management decision making and make a more defensible case for their annual budget requests; and

· In the annual performance appraisals of managers, give significant weight to how effectively their organizational unit has met its policy and program objectives as measured by the performance indicators.

Box 1. Lessons from the UK on Managerial Accountability

UK experience with implementing such reforms – usually branded under the umbrella of “New Public Management” – offers cause for optimism. Holding managers accountable for the delivery of outputs and outcomes, accompanied by allowing more discretion over the control of inputs (e.g. staff and finance), has improved the performance of public institutions in service delivery and led to greater value for money. Designing appropriate performance indicators and robust and reliable monitoring and evaluation systems are key to holding managers properly accountable. The Cyprus public service lags behind most OECD countries’ public administrations in applying this management practice, but, with targeted capacity building, it could catch up very quickly.

Source: Government of the United Kingdom

In addition, managers could be provided with training to enhance their performance management skills.[footnoteRef:17] This training, could include: [17: Managers in the Cyprus Public Service are currently receiving training within the framework of a project co-financed by the European Social Fund for strengthening the strategic, leadership, and management capacity of the Public Administration.]

· Setting SMART goals.[footnoteRef:18] [18: SMART stands for Specific, Measurable, Attainable, Relevant and Time-bound goals.]

· Tracking both organizational and individual staff performance.

· Techniques for motivating staff and encouraging staff learning and professional growth.

· Techniques to deal with under-performing staff, including how to pursue staff removal when other options (such as continuous feedback, encouragement, training, mobility) are not sufficient to deal with poor performance.

These reforms will focus attention on efficiency and effectiveness. They will also create a framework for organizational accountability since progress towards the achievement of objectives can be monitored and assessed by policy makers, the legislature, and the public.

Complementary Reform Options for Greater Effectiveness

Reduce disincentives for staff mobility across the public administration. A recent IMF report clarifies some of the more important constraints to staff mobility in Cyprus:

“Legal constraints on mobility across and within government entities impede the reallocation of employees to higher priority activities. Only limited categories of workers can be transferred across and within ministries and departments of the government. These include some support workers (including general clerical/administrative staff, cleaners, messengers who are contractually “interchangeable”), and entry level employees (through “duty assignment”). Recent laws have allowed these workers to be reallocated within ministries without central-level approval. However, reallocations across ministries and departments still require the approval of the Public Administration and Personnel Department of the Ministry of Finance (PAPD). In addition, severe limitations exist in reallocating non-interchangeable staff (including hourly-paid workers and technical staff). For the remainder of the civil service, the only option available currently is through temporary secondments.”[footnoteRef:19] [19: IMF (2013). Since the IMF report was written, legal changes have also created the possibility of duty-assignment, which enables Permanent Secretaries of line Ministries to assign-duties to permanent non-interchangeable staff (at the entry level) serving in any departments/services within the Ministry, to another department/services within the Ministry in order to meet service requirements/shortages. Another recent legal amendment has allowed certain casual staff serving in Ministries/Departments/Services to be turned into interchangeable staff, thereby increasing the number of interchangeable staff to 28% of the total number of employees; i.e., slightly more than one-quarter of employees can be reallocated internally and across Ministries/Departments at any time. To date, these devices are only infrequently used.]

Recent changes to the legislative framework regarding secondments have eliminated the previously required clearance by the affected staff member.[footnoteRef:20] While this change has removed a significant rigidity in the legal framework, additional barriers, such as unduly restrictive schemes of services, continue to restrict mobility. At the same time, there are no career-enhancing incentives for individuals to move around the public administration. Many of the changes we propose (e.g. increasing open competitions for post, stricter performance management, removing the combined salary scales, limiting the number of annual increments, and reducing the numbers of schemes of service) should, over time, change perceptions to regard mobility as both positive and career- enhancing. For many promotional posts, experience gained outside of the Department should also be made either a qualifying condition or added advantage for promotion. [20: Those recent changes have also eliminated the previous restrictions on the duration of secondments.]

Reduce or eliminate the constraint on the dismissal of civil servants during their initial two-year probation period. Existing laws place all new recruits into the civil service on a two-year probation period, where their performance is appraised every six months. However, they cannot be dismissed until the end of the two-year probation period. At a minimum, the non-dismissal constraint should elapse after two performance appraisals (i.e. the end of the first year of the two-year probation period), which would give management needed flexibility to remove underperforming recruits in a more timely fashion.

This change would not, however, be sufficient, as the PSC has demonstrated little inclination to exercise its discretion to dismiss under-performing civil servants. Without a greater willingness (or ability) to sanction under-performing staff, it will be difficult to achieve a more performance-oriented civil service. To avoid the potential abuses to authority, however, it will not be possible to discipline under-performing staff until performance accountability mechanisms have been adequately improved. In the interim, more training should be provided to the PSC staff and commissioners on the importance of addressing this issue. At the same time, managers, in the new strategic framework context, should be given greater authority to dismiss or replace under-performing staff in order to build high-performing teams capable of achieving their targets.

Reform Options for the Performance Appraisal Process

To ensure that merit and performance are the key determinants of career prospects, including both promotions and salary growth, the annual personnel performance appraisal process needs to be amended. The current proposal from the Public Administration and Personnel Department (PAPD) is a welcome initiative and, encouragingly, builds in a staff development component for managers to provide more continuous feedback on performance. It is, however, missing one important feature. Currently, there is no device to discipline the pattern of ratings assigned to staff, either at the ministry or agency level. This recommendation was initially included in an earlier proposal, but was subsequently dropped due to opposition from various stakeholders. To ensure a well-functioning performance appraisal process, the following reforms should be considered:

· Staff contributions to policy and program objectives are reflected:

· Increase weight of staff contribution to organizational targets in annual performance review process: Make staff contributions to the performance targets of the organizational unit within which they are posted a core criterion in their annual performance appraisal.

· Negotiate annual performance targets for all staff, reflecting the ways in which their work contributes to the policy & program objectives of the organizational unit within which they are posted.

· Require evidence on performance relative to targets: Require evidence of the extent to which staff performance targets have been met as an input to their annual performance appraisal.

· Discipline annual personnel performance appraisal process: Introduce mechanisms to discipline the annual personnel performance appraisal process to ensure variance in ratings. The simplest option is the widely employed practice of setting quotas (either minimums or maximums) on the proportion of performance ratings falling in particular categories. Quotas, however, should only be applied at the ministry or agency level to ensure that there is a sufficient number of staff to generate a statistically reliable distribution. In a system with five performance categories, it is typical for the disciplining mechanism to limit ratings in the top category to no more than 5%, in the second to top category to 15%, and to require at least 1% in the lowest rating and 5-10% in the second lowest rating category. These parameters would allow around 70-75% of staff to be rated in the middle category. Names for the categories are sometimes couched in relatively superlative terms to lessen a feeling of being underappreciated among the majority of staff who receive the middle rating. For example, 1=Superior; 2=Excellent; 3=Good; 4=Satisfactory; and 5=Unsatisfactory. While this disciplining mechanism is rarely popular among staff (since most receive a rating that they believe is lower than what they deserve), and has already been a subject of controversy when it was first proposed in Cyprus, it is critical if the annual performance review process is to be credible.

Box 2. Lessons from the UK on Performance Appraisal

In the UK, for example, assessment criteria are derived from a combination of common competencies (by level), describing necessary skills and behaviors alongside role specific objectives linked to organizational targets. A quota system (25:70:5) is used at the agency and departmental level in order to ensure variance in ratings. The approach known as “360 degree” feedback is also being introduced to enhance the fairness and reliability of performance ratings (see Annex 2 for additional lessons on performance appraisal from other EU member states).

Source: Government of the United Kingdom

· Institutionalize checks on the performance ratings to ensure fairness and reliability:

· Require input on each staff member’s performance from multiple parties: It is important to gather feedback from both inside and outside an employee's immediate work circle. Input can be provided by (i) colleagues and peers; (ii) clients (i.e., persons intended to benefit from the work undertaken by the staff member); (iii) leaders of teams on which the staff member works. This approach is formally known as a 360° appraisal.

· Require supervisors to discuss the appraisal with staff before it is finalized. This consultation is important to ensure that the staff member is afforded an opportunity to contest any ratings they feel do not accurately reflect their performance.

· Require all supervisors within a large organizational unit (e.g., a ministry or agency) to discuss all their appraisals with peers before finalizing them. This collaboration across the institution is important to ensure comparable standards within a larger organizational unit.

· Provide an independent administrative redress mechanism through which aggrieved staff can contest their performance ratings if they believe they seriously misrepresent their performance. This mechanism would allow staff to challenge important HRM actions without necessarily taking their grievances to a full judicial hearing. The (more burdensome) Supreme Court’s involvement could thereby be limited to the more difficult to resolve cases.

At the same time, special consideration must be given to the performance appraisal of interchangeable staff. Although these employees work across all government ministries and agencies, they are formally mapped to the PAPD of the MoF. In the context of a strengthened performance appraisal system, however, where staff receive more constant feedback on their performance from their manager, the Government could consider re-mapping interchangeable staff from PAPD directly to their current department. Under this scenario, the performance appraisal could then be undertaken by their direct manager.

NSGI and GoC are in the process of completing a new performance appraisal system. The preceding analysis in this section is preliminary and complementary to the ongoing efforts, which is expected to be finalized by the end of June 2014.

Strengthening Performance in Promotion Decisions[footnoteRef:21] [21: See Annex 3 for a discussion of career-based versus position-based HRM systems.]

Reforms to the annual performance appraisal process would need to be complemented with significant reforms to strengthen the roles of performance and merit in promotions processes. These reforms include:

· Reduce or eliminate seniority as a criterion for promotion: Options include:

· Drop seniority as a criterion.

· Include seniority only as a minimum requirement, but not as an element in the score for determining promotions decisions.

· Leave seniority as a full criterion, but reduce its weight from one-third to 10%.

· Substitute “merit” for “qualifications” as a criterion for promotion, or redefine “qualifications” to mean:

· Not only academic credentials, but the quality and relevance of those credentials for the promotion position.

· Relevant experience, regardless of whether it has been obtained inside or outside the Cyprus public administration.

· Formal testing of job-relevant knowledge and skills for some positions.

· For managerial positions, testing of “soft” skills by trained and experienced professionals for teamwork, collaboration, communication, leadership, etc.

· Assess performance more comprehensively: In particular, include not only recent annual personnel performance appraisal scores, but also additional inputs, such as:

· Feedback from peers, clients and supervisors of candidates.

In addition to strengthening the roles of merit and performance in promotions decisions and salary growth, a number of steps should also be undertaken to enhance the contestability of promotions decisions:

· Enhance competition in recruitment and promotions decisions

· Open competition: Make open competition (open to both internal and external candidates) the default requirement for filling positions, regardless of whether the selection process is for “entry and promotion” or “promotion”. Restrictions to internal competition should only be allowed under special circumstances (with clearly specified criteria for what constitutes “special circumstances”).

· Revise schemes of service to permit greater competition: Undertake a fundamental review of the schemes of service (a significant barrier to greater mobility) to significantly reduce their number. They are currently so numerous and tightly drawn that the eligible field is often by definition confined to the particular department, which generates a significant barrier to greater mobility.

· Employ a three-stage decision process for all promotions decisions to ensure that no single agent is likely to control the outcome of a promotion or hiring decision: Redesign both the “promotions” and the “promotions and entry” processes to reduce the odds that any single agent is able to control the outcomes at each stage of that procedure.

Pay Policies Not Linked to Performance

Key Findings

· Significant pay increases independent of performance

· Public sector pay significantly higher than private sector

Proposals

· Reform combined pay scales

· Restrict “annual increments”

· Selective salary freeze

· Subject CoLA to budgetary constraints

Pay reform must be linked to broader HRM reforms. The pay reforms suggested in this section are directed at two central objectives: establishing salary structures designed to be supportive of HRM practices that reward merit and performance primarily through the promotions process and supporting a fiscally sustainable wage bill. If the salary structure and its dynamics are not also reformed in ways that ensure that merit, performance and promotions become the primary path to career salary growth the above HRM reforms will likely create as many problems as they solve. In particular, they may not be affordable if the current salary scales are maintained. Integrating the above HRM reforms with a new salary structure will help to ameliorate possible staff dissatisfaction with their reduced career salary growth prospects that are unrelated to performance and promotion.

Wage increases have been driven by institutional rules rather than performance. The growth in salaries is driven by a combination of g, compound increases, and CoLA, which accrue to every civil servant each year with weak linkages to performance (Figure 1). From 2007 to 2012, CoLA was the largest source of growth in total remuneration followed by general increases. Increases to base salaries through promotions and annual increments were the smallest contribution to wage increases over the period. Once the crisis hit, these wage growth parameters were among the main sources of concern regarding the sustainability of the wage bill.

Figure 1. Contributions to Earnings per Employee Growth (Percentage points)

Notes: The series reflect average earnings of the permanent and casual employees working in the central government and cover five components: Basic salary including annual increment, general increase, COLA, overtime, and allowances. Earnings per employee are calculated as total earnings of the permanent and casual employees in a given year divided by average monthly permanent and casual employment in the same year. The annual growth in earnings per employee is decomposed into its five components. The contribution of each component is calculated as annual growth of earnings from that component (by employee) weighted by the component's share in total earnings per employee (see Box A1 in Annex 1 for a description of the methodology). Source: World Bank calculations based on Ministry of Finance personnel database.

Although the CoLA formula has been revised under the Troika program, it should still be subject to budgetary constraints. Until recent crisis measures were introduced, the gross earnings of workers were revised every six months on the basis of the consumer price index percentage of the preceding six month period. Under this mechanism, wage increases accrued independently of productivity gains and grew faster than the actual inflation rate in the years leading up to the crisis. CoLA payments are currently suspended until the end of March 2016 and the formula has also been recently revised under the Troika program for future payments (see Box 3). These changes are welcome, but an additional component that subjects CoLA payments to budgetary constraints should be included. This element would grant budgetary officials extra flexibility in times of fiscal stress.

Box 3. Recent Changes to the CoLA Formula

· A lower frequency of adjustment, with the base period for calculating CoLA being lengthened from the current period of six months to twelve months. Indexation would take place on 1st January each year;

· A mechanism for automatic suspension of application and derogation procedures during adverse economic conditions, such that if in the second and third quarters of a given year negative rates of growth of seasonally adjusted real GDP are registered, no indexation would be effected for the following year; and

· A move from full to partial indexation, with the rate of wage indexation being set at 50% of the rate of increase of the underlying price index over the previous year.

Regularized institutional pay increases are magnified by “combined” salary scales. While there are 16 distinct salary scales, plus a set of “Fixed Salary” scales, for setting individual salaries within Cyprus’ public administration, several of those scales have been “combined” into single “combined salary scales” (see Figure 2). All salary scales allow public servants to move up the steps within the grade defined by that scale each year without having to be promoted, as long as they have received at least a satisfactory annual performance evaluation. Most salary scales include between 11 and 13 steps (with a few including as few as 6 or as many as 15). A “combined salary scale”, on the other hand, allows a public servant governed by that “combined salary scale” to continue this progression across all the salary scales included within that “combined salary scale”. This pay scheme significantly increases the number of steps (pay increases) a public servant can ascend over the course of his/her career without ever being promoted.

Figure 2. Permanent and Casual Wage Bill by Scale, 2012

Note: The other category (comprises the remaining scales, including fixes salaries, which have a very small share in total employment – approximately 2 percent).

Source: World Bank calculations based on Ministry of Finance data.

Combined salary scales guarantees automatic wage increases for employees even if they are not promoted for virtually an entire career (18-29 years). Table 3 illustrates the salary growth potential in each of three major “combined” salary scales. The A2-5-7(ii) combined salary scale (clerical staff) allows an employee to obtain annual salary increments for up to 29 years, which would generate salary growth of 125 percent over that period, or 18,856 euros. The A8-10-11 combined scale (staff with university degrees, such as Administrative Officers, Economic Officers, etc.) affords the potential for 21 years of salary increases without promotion and a potential salary increase of 119 percent or 29,139 euros. Similarly, the A9-11-12 combined scale (lawyers, engineers, accountants) allows a civil servant to increase their salary over an 18 year period by 96 percent, or 29,283 euros, without promotion and on the basis of seniority rather than performance.

Table 3: Combined Salary Scales Parameters[footnoteRef:22] [22: Sources: “Salary Scales 2012-2015” file and “Main Career Structures” document provided by the Public Administration and Personnel Department (PAPD), Ministry of Finance, Government of Cyprus (September 2013).]

Salary Growth Potential Parameter

Combined Salary Scale

A2-5-7

A8-10-11

A9-11-12

Entry Level Total Salary (Euros)

15,109

24,498

30,409

Salary Growth Potential without Promotion

Maximum Total Salary Possible (Euros)

33,965

53,637

59,691

Total Possible Salary Growth (Euros)

18,856

29,139

29,283

Total Possible Salary Growth (%)

125

119

96

Maximum Number of Step Increases Without Promotion

29

21

18

Source: Government of Cyprus

Combined pay scales reduce the value of promotions. With combined pay scales, public servants who are not promoted receive nearly the same compensation as staff who are promoted. Figure 3 shows how the average compensation changed for a sample of employees who were active in 2012 and who had been working since at least 2002. Within the sample, nearly 40 percent of the staff had not been promoted (and remained in the same pay scale) had an average pay increase of 90 percent over this time period, slightly less than the average pay increase of 105 percent for staff who had been promoted. If the “non-promoted” sample is further restricted to non-promoted staff in the A8-A10-A11 combined scale, they received an even higher salary increase of 96 percent in this period.

Figure 3. Effects of Seniority on Earnings of Permanent/Casual Employees (in Euros)

Notes: The sample is restricted to employees who worked for the entire 2002-12 period. When an employee changed his scale he is assumed to be “promoted.”

Source: World Bank calculations based on Ministry of Finance data

Combined pay scales significantly distort incentives in the public sector. This policy contributes to demotivated staff, excessively high salaries, particularly late in staffs’ careers, and a very weak relationship between the human capital requirements of positions, including their levels of responsibility and remuneration. Another consequence of the combined pay scales is a gap between public and private sector pay. Pashardes (2011a) finds that public sector remuneration exceeded private sector remuneration by 43% in 2009 (not controlling for career stream, education, and experience). Controlling for career stream, education and experience of jobholders, Pashardes (2011a) finds that public sector total remuneration in Cyprus exceeds that earned by similar workers in the private sector by an average of 27% (Table 4).[footnoteRef:23] By occupational category the findings suggest: [23: Analysis is based on individual household data from the Cyprus Statistical Service household survey. ]

· Only managers and senior officials are actually paid less than their private sector counterparts.

· Teachers are the most significantly overpaid staff, receiving more than double what their counterparts earn as private sector teachers in Cyprus.[footnoteRef:24] [24: A large body of empirical literature finds a tendency for lower skilled workers, as well as women, to receive wage premiums for working in the public sector, while higher skilled workers and males tend to receive lower salaries in the public administration; so this finding is consistent with that empirical literature. The magnitudes of these Cyprus premiums, however, are considerably larger than typically found in this empirical literature, particularly for teachers and low-skilled workers. To illustrate, teachers, for instance, are found in a recent study by Giordano et al. (2011) to earn between about 5% and 34% more in these public sectors, while those in Cyprus earn more than 100% more than their private sector counterparts, according to the Pashardes study.  Similarly, low-skilled workers in these studies earn premiums in the 8% to 30% range, while the Pashardes study found premiums of around 79% for unskilled workers and 57% for service workers. ]

· Clerical staff, unskilled workers, and service and sales workers are the next highest paid group relative to comparable private sector workers, receiving salary premiums of 31 %, 57% and 79% respectively, relative to comparably educated and experienced workers in those same occupations/career streams.[footnoteRef:25] [25: Pashardes (2011).]

· The remainder of the public administration workforce receives pay premiums in order of 15 to 30 percent.[footnoteRef:26] [26: Pashardes (2011b) extends the Pashardes (2011a) analysis, examining public-private pay differentials at three separate times – 1996, 2003, and 2009 – and finds important changes over this period. In particular, the competitiveness of salaries for (a) managers and senior administrators declined by roughly half from 152% to 82%; (b) “qualified personnel” (i.e., those with college degrees) declined from 146% to 106%; (c) teachers rose dramatically from 128% to 228%; and (d) unskilled laborers increased from 128% to 166%. Changes in competitiveness among the remaining personnel categories have occurred as well, but have been more modest. The small differences in the 2009 estimates between the two Pashardes studies reflect slightly different samples as the (2011b) study used matched samples to ensure comparability of the estimates across years]

In another study that compares the public-private pay gap across all EU members (excluding France and Malta but including Iceland and Norway), Cyprus is second only to Luxembourg in terms of the premium afforded to public sector employees over private sector employees with comparable levels of education and experience.[footnoteRef:27] Moreover, these estimates almost certainly underestimate the premium for public service employment, particularly in Cyprus, since (i) the health and education sectors are not included and (ii) they do not control for non-remuneration terms and conditions of employment that typically make public employment more attractive, such as the greater due process protections and, in Cyprus’ case, the especially generous pension benefits of public sector employment. Given the other research on Cyprus,[footnoteRef:28] it is likely that the premium offered for employment in the public sector is among the highest within the EU, if not the largest. [27: Christofides and Michael (2013): p. 22, Figure 13.] [28: Pashardes (2011a, 2011b).]

Table 4: Public-Private Pay Differentials by Profession

Occupational Group/Career Stream

Public remuneration as a % of private sector remuneration

Managers, Senior Officials

80

Teachers

207

Other Professionals

113

Technicians, Associates

121

Clerks, Cashiers

131

Service and Sales Workers

157

Skilled Workers

118

Unskilled Workers

179

Average

127

Source: Pashardes (2011a)

In addition to these public sector pay premia, Cyprus’ public sector pension provisions also significantly enhance the attractiveness of lengthy public sector employment.[footnoteRef:29] At least two provisions of Cyprus’ public sector pensions policies and practices provide significant additional benefits to public servants: [29: See Annex 6 for a review of working conditions in the Cyprus public sector.]

· A unique one-off retirement bonus (lump sum benefit) between 4.7 and 5.2 times the annual pension is provided to all eligible public servants.[footnoteRef:30] [30: IMF (2011), p. 12.]

· Public servants receive pensions under both the General Social Insurance Scheme (GSIS), as well as the Government Employees Pension Scheme (GEPS). If a public servant leaves the public service voluntarily before the age of 45, they are not entitled to any pension benefits under the GEPS.[footnoteRef:31] Early retirement can be drawn at the age of 55 years (or 58 for entrants into public service after 1st July 2005) without any actuarial reduction of benefits.[footnoteRef:32] Thus, the generous GEPS benefits, at least up to the age of 45, create a strong incentive for public servants not leave public service. [31: Article 27, Pensions Act 1997.] [32: Petmesidou (2012).]

· However, recent changes to the pension system may act to increase low turnover rates in the public sector. Public servants recruited after October 2011 will only be eligible for GSIS, rather than both GEPS and GSIS. All public servants will continue to be eligible for the significantly more generous GSIS, which is afforded to all formal sector employees within the country, be they in the public or private sector.

The high pay differential between public and private pay, coupled with the generous pension provisions, contributes to a very low turnover rate in the public administration.[footnoteRef:33] The turnover rate has averaged 4.6 percent over the last decade, significantly lower than average rates in other OECD countries, although this number increased slightly in the last two years (Figure 4). A study on the United Kingdom’s public sector, for example, found that the average overall turnover rate was 8.2 percent in 2011,[footnoteRef:34] while other research corroborates these findings, finding turnover rates ranging between 5 percent and 30 percent across departments in the United Kingdom’s public administration.[footnoteRef:35] Similarly, research in the United States found average turnover rates within the state and local public administration of 9 percent in 2011.[footnoteRef:36] In Cyprus, new hires have exceeded departures from 2002 to 2009, reflecting the growth in the size of the public workforce. This trend reversed, however, in 2010 when the Government reduced hiring (and additional expenditures) during the crisis period. [33: Due process protections also contribute to low public sector turnover rates, but would apply to any due-process-protected civil service. Compared to other countries’ public administrations suggest, the lower turnover rates within Cyprus’ public administration suggest that pay and pension are the key factors driving this low rate.] [34: Chamberlain (2011).] [35: Institute for Government (2011).] [36: Jacobs (2011).]

Figure 4. Central Government Turnover Rates (% of average monthly employment)

Source: World Bank calculations based on Ministry of Finance data

Reform combined salary scales

The combined salary scales, in effect, guarantee salary growth over a significant portion of a public servant’s career regardless of their performance. Although they were initially adopted as a response to poor career prospects for many civil servants, they led to a significant increase in salaries, which undermines the fiscal sustainability of the wage bill, and reflect seniority rather than performance. To address these twin problems, it is essential to make performance, rather than seniority, the primary determinant of salary growth over a public servant’s career. This change requires two sets of actions: (i) make performance-based promotions the primary means of achieving salary increases over the course of one’s public service career, and (ii) limit salary growth potential among public servants who are not promoted. Options for addressing (i) are addressed in the next section of this report, while two options are proposed for addressing (ii): (a) compress each combined scale, or (b) eliminate all combined salary scales.

Option 1: Compress each combined scale, by reducing the total number of steps in each combined salary scale to around seven, while simultaneously ensuring that annual increments are small. The following parameters could guide this salary reform option:

· Scale anchoring: Each new combined scale would be anchored to the current lowest (entry level) salary included within that scale.

· Limited steps: Each new combined scale would include a small number of steps; say, five to seven.

· Within-scale annual salary increases: Each step would afford a small annual increase in total remuneration.

· Transition provisions: Each existing public servant would be assigned to the highest step in the revised scale that is at or below their current remuneration level. At the same time, existing staff whose pre-reform remuneration exceeded their post-reform mandated remuneration would have their remuneration frozen until either (i) the revised salary scale “catches up” with their existing remuneration or (ii) they are promoted.

· New recruits: New recruits into the post-reform salary scale would have their remuneration determined by the post-reform salary scale.

Option 2: Eliminate combined salary scales. The following parameters could guide this salary reform option:

· Combined scales eliminated and replaced: Each combined salary scale would be broken into its component non-combined elements; e.g., the combined A2-A5-A7(ii) would be replaced by three separate scales: A2, A5, and A7(ii). This separation would necessitate new legislation to change the provisions within existing schemes of service and the creation of new schemes of service to define new promotion posts. These promotion posts should derive from a staffing review that is linked to the organizational and unit level objectives and performance targets and should not be undertaken in isolation. The budgetary and strategic management reforms will also provide an opportunity to create new promotion posts, subject to budget constraints.[footnoteRef:37] [37: Creation of promotions posts through restructuring is not an available option under the current employment freeze; but that is a temporary constraint.]

· Scale anchoring: Each non-combined scale would be anchored to the current lowest step in each of the pre-reform A2, A5, and A7(ii) scales.

· Limited steps: Each non-combined scale would include a small number of steps; say, five to seven.

· Within-scale annual salary increases: Each step would afford a small annual increase in total remuneration.

· Transition provisions: Existing staff within a combined scale would be subject to a detailed review of their qualifications and performance, and would then be assigned to the non-combined scale for which their qualifications and performance are most appropriate (to be decided by the Civil Service Commission, PAPD, and the Minister of Finance). In addition, each public servant assigned to a particular new non-combined scale would be assigned to the highest step at or below their current remuneration level.

· New recruits: New recruits into the post-reform salary scale would have their remuneration determined by the post-reform salary scale.

The elimination of combined salary scales would not eliminate all salary increases for non-promoted staff. They would still receive the annual salary increments embedded in their non-combined salary scale as well as the CoLA adjustment. Staff at the top of their salary scales, however, would no longer receive an annual increment. This policy was also applied in the United Kingdom, where, after a reasonable time period (5 to 10 years), the highest increment is reached and the concept of a “rate for job” is applied.

Reform “annual increments”

Restricting the number and amount of annual increments would ensure that promotions based primarily, though not exclusively, on performance would be the primary means of salary progression over one’s career. We offer three distinct options to ensure ensuring that annual increments remain modest: (i) keep them small and independent from performance; (ii) allowing modest performance-based one-off bonuses for a limited number of staff; and (iii) consolidating them with CoLa and general increases.

Option 1: Limit number and value of annual increments. To ensure that all salary scales permit only modest salary growth independent of performance-based promotions, we propose: (i) a limited number of steps in each scale, for example, five to seven; and (ii) modest increases in each annual increment, which would be independent of performance. International experience suggests that the range within most grades should be between 20 to 50 percent. Currently, several pay scales in Cypriot pay structure are above this range and should thus be curtailed (see Figure 5). The new ranges should also be conditioned to the expected pay increases in the private sector for similar positions and subject to budget constraints. At the same time, the annual increments should be restricted to less than 4 percent of base salary.

Figure 5. Salary Range and Average Annual Increment by Scale, 2014

Note: Salary range is defined as the percentage difference between the minimum and maximum in each scale.

Source: Ministry of Finance

Option 2: Modest performance-related one-off bonuses: The Government could also reward high performing staff with a one-off bonus. High performing staff could be defined as those whose annual performance assessments were in the top x% of all performance ratings, where the x% parameter would need to be set in a Government policy. This reform will not work, however, if the existing annual personnel performance assessment practices persist since virtually all staff receive the highest possible rating. Accordingly, this reform should not be undertaken until the annual personnel performance assessment process has been revised so that it generates a distribution that adequately reflects different levels of performance. Annex 4 provides a comprehensive review of international empirical evidence on the experience with efforts to link pay to performance. A key finding of that review is that pay-for-performance (P4P) has been shown to improve performance outside the core public sector, but only in cases where it is possible to reliably measure and weight all important dimensions of performance. The evidence on P4P within most core public administration functions suggests that it is quite difficult to get P4P to deliver on its performance-enhancing promises within the public administration and other areas where outputs are difficult to objectively measure.[footnoteRef:38] [38: Anderson, Reid and Ryterman (2003); Beer and Cannon (2004); Behn (2004); Cardona (2007); French (2005); Hood and Dixon (2010); Houston (2009); Hutton (2010); Kellough and Lu (1993); Kellough and Nigro (2002); Kernaghan (2011); Kerr (1975); Ketelaar, Manning and Turkisch (2007); Kim (2002); Marsden (2004, 2009); Mayer and Davis (1999); McCourt (2006); Milkovich and Wigdor (1991); Moynihan (2008); Moynihan and Pandey (2007); Murnane and Cohen (1986); Nagin, et al (2002); O’Donnell and O’Brien (2000); OECD (2005b, 2005c); Perry, Engbers and Jun (2009); Perry and Paarlberg (2006); Recanatini, Prati and Tabellini (2005); Reid (2006, 2007, 2010); Schick (1998); Straberg (2010); UK National Audit Office (2008); Weibel and Osterloh (2009); White (2000); Willems, Janvier and Hendrickx (2006).]

Option 3: Combine annual increments with CoLA: The annual pay increase approved as part of the annual budget process would revise pay levels in accordance with (a) fiscal sustainability; (b) competitiveness of remuneration relative to the domestic private sector; and (c) CoLA. This option implies that everyone would get a pay increase whenever those three considerations yield an increase in the wage bill. The amount